Cargando…

MHA-Net: A Multibranch Hybrid Attention Network for Medical Image Segmentation

The robust segmentation of organs from the medical image is the key technique in medical image analysis for disease diagnosis. U-Net is a robust structure for medical image segmentation. However, U-Net adopts consecutive downsampling encoders to capture multiscale features, resulting in the loss of...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhang, Meifang, Sun, Qi, Cai, Fanggang, Yang, Changcai
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Hindawi 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9560845/
https://www.ncbi.nlm.nih.gov/pubmed/36245836
http://dx.doi.org/10.1155/2022/8375981
_version_ 1784807842648162304
author Zhang, Meifang
Sun, Qi
Cai, Fanggang
Yang, Changcai
author_facet Zhang, Meifang
Sun, Qi
Cai, Fanggang
Yang, Changcai
author_sort Zhang, Meifang
collection PubMed
description The robust segmentation of organs from the medical image is the key technique in medical image analysis for disease diagnosis. U-Net is a robust structure for medical image segmentation. However, U-Net adopts consecutive downsampling encoders to capture multiscale features, resulting in the loss of contextual information and insufficient recovery of high-level semantic features. In this paper, we present a new multibranch hybrid attention network (MHA-Net) to capture more contextual information and high-level semantic features. The main idea of our proposed MHA-Net is to use the multibranch hybrid attention feature decoder to recover more high-level semantic features. The lightweight pyramid split attention (PSA) module is used to connect the encoder and decoder subnetwork to obtain a richer multiscale feature map. We compare the proposed MHA-Net to state-of-art approaches on the DRIVE dataset, the fluoroscopic roentgenographic stereophotogrammetric analysis X-ray dataset, and the polyp dataset. The experimental results on different modal images reveal that our proposed MHA-Net provides better segmentation results than other segmentation approaches.
format Online
Article
Text
id pubmed-9560845
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher Hindawi
record_format MEDLINE/PubMed
spelling pubmed-95608452022-10-14 MHA-Net: A Multibranch Hybrid Attention Network for Medical Image Segmentation Zhang, Meifang Sun, Qi Cai, Fanggang Yang, Changcai Comput Math Methods Med Research Article The robust segmentation of organs from the medical image is the key technique in medical image analysis for disease diagnosis. U-Net is a robust structure for medical image segmentation. However, U-Net adopts consecutive downsampling encoders to capture multiscale features, resulting in the loss of contextual information and insufficient recovery of high-level semantic features. In this paper, we present a new multibranch hybrid attention network (MHA-Net) to capture more contextual information and high-level semantic features. The main idea of our proposed MHA-Net is to use the multibranch hybrid attention feature decoder to recover more high-level semantic features. The lightweight pyramid split attention (PSA) module is used to connect the encoder and decoder subnetwork to obtain a richer multiscale feature map. We compare the proposed MHA-Net to state-of-art approaches on the DRIVE dataset, the fluoroscopic roentgenographic stereophotogrammetric analysis X-ray dataset, and the polyp dataset. The experimental results on different modal images reveal that our proposed MHA-Net provides better segmentation results than other segmentation approaches. Hindawi 2022-10-06 /pmc/articles/PMC9560845/ /pubmed/36245836 http://dx.doi.org/10.1155/2022/8375981 Text en Copyright © 2022 Meifang Zhang et al. https://creativecommons.org/licenses/by/4.0/This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
spellingShingle Research Article
Zhang, Meifang
Sun, Qi
Cai, Fanggang
Yang, Changcai
MHA-Net: A Multibranch Hybrid Attention Network for Medical Image Segmentation
title MHA-Net: A Multibranch Hybrid Attention Network for Medical Image Segmentation
title_full MHA-Net: A Multibranch Hybrid Attention Network for Medical Image Segmentation
title_fullStr MHA-Net: A Multibranch Hybrid Attention Network for Medical Image Segmentation
title_full_unstemmed MHA-Net: A Multibranch Hybrid Attention Network for Medical Image Segmentation
title_short MHA-Net: A Multibranch Hybrid Attention Network for Medical Image Segmentation
title_sort mha-net: a multibranch hybrid attention network for medical image segmentation
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9560845/
https://www.ncbi.nlm.nih.gov/pubmed/36245836
http://dx.doi.org/10.1155/2022/8375981
work_keys_str_mv AT zhangmeifang mhanetamultibranchhybridattentionnetworkformedicalimagesegmentation
AT sunqi mhanetamultibranchhybridattentionnetworkformedicalimagesegmentation
AT caifanggang mhanetamultibranchhybridattentionnetworkformedicalimagesegmentation
AT yangchangcai mhanetamultibranchhybridattentionnetworkformedicalimagesegmentation